July 2010
No.9
Unilateral measures and emissions mitigation
Shurojit Chatterji (Singapore Management University), Sayantan Ghosal (University of Warwick),
Sean Walsh (Centre for International Governance Innovation) and John Whalley (Centre for
International Governance Innovation
and University of Western Ontario and NBER)
WORKING PAPER SERIES
Centre for Competitive Advantage in the Global Economy
Department of Economics
Unilateral measures and emissions mitigation
Shurojit Chatterji , Sayantan Ghosal , Sean Walsh# and John Whalley##
Singapore Management University,
University of Warwick,
#
Center for International Governance Innovation,
##
Center for International Governance Innovation
and University of WesternOntario and NBER
June, 2010
Abstract
Do unilateral measures to cut emissions provide an adequate foundation for
global climate change negotiations from a post-Copenhagen perspective? We document the extent and variety for unilateral measures. In a formal model, we examine
the conditions under which global learning, by building on the positive spillovers
generated by unilateral measures, delivers cumulative emissions reduction over time.
Using our results, we analyze the key features of a global policy regime builds on
unilateral measures to accelerate convergence to a low carbon world.
JEL Classi…cation: Q54, F53, Q55, O33.
Keywords: Unilateral initiatives, mitigation, global learning, technology transfer.
We would like to thank participants ESRC funded Climate Change Workshops at Warwick in 2009
and 2010 for their helpful comments. E-mail addresses: shurojitc@smu.edu.sg; S.Ghosal@warwick.ac.uk;
swalsh@cigionline.org; jwhalley@uwo.ca.
1. Introduction
The Copenhagen Accords1 , while non-binding, sets out the foundation of a framework
for the unilateral actions of all the parties relating to mitigation and adaptation related
activities. Do such unilateral measures provide an adequate foundation for global climate
change negotiations from a post-Copenhagen perspective?
This paper studies the interplay between unilateral measures and multilateral negotiation in relation to global climate change mitigation from a post-Copenhagen perspective.
We, …rst, document the extent and variety of existing unilateral measures and in a model
of global learning, examine the conditions under which global negotiations, building on
the positive spillovers generated by such unilateral initiatives result in cumulative emissions cuts over time. Based on our results, we develop a policy proposal that addresses
the obstacles if further global negotiations, building on the Copenhagen Accords, have
the potential to achieve the goal of a 2 C cap on temperature rise2 .
We note that participation and compliance in a broad-based multilateral initiative that
aims to go further than what individual countries are unilaterally willing to commit may
be hard to achieve. Any broad based global cooperation that requires nations to commit
to emission cuts beyond what nations are unilaterally willing to undertake is unlikely to
be stable (i.e. immune to deviations by a nation or a coalition of nations). By delaying
participation or by not complying with a global agreement to cut emissions, a deviating
country or a coalition of such countries can continue capture the short-term bene…ts from
continuing with high carbon economic activities but pass a signi…cant portion of the costs
to others (other countries, future generations): free riding in the presence of negative
externalities limits the scope of multilateral cooperation in the presence of weak property
rights (Shapley and Shubik (1969) and Starrett (1973) were among the …rst to make this
point).
Although multilateral cooperation has been a hard to achieve, a number of unilateral initiatives are already underway to cut emissions. We begin by documenting the
extent, form and variety, some at a national level but mainly at a subnational (regional/urban/individual) level, of existing unilateral measures. Examples of such unilateral measures include community based programs such as free bicycle plans, such as
in Copenhagen, Denmark, or watershed renewal programmes, such as in rural communi1
The text that came out of the Copenhagen meeting hosted by the United Nations Framework Convention on Climate Change (UNFCCC) in December 2009 is referred to as the Coppenhagen Accords.
2
As set out in the reports of the IPCC. A more ambitious target 1.5 C cap has been recently suggested
by the low-lying island states in Copenhagen.
2
ties bordering major rain forests as in Mexico. As such unilateral measures consist of a
mix of local adaptation together with increasing capacity in renewables and the reduction of energy consumption either directly and indirectly (via energy saving techniques,
infrastructure renewal and increased e¢ ciency).
If a city introduces a raft of measures to induce more use of public transport and
some of these measures are successful3 , other cities elsewhere in the world can learn from
its experience and implement similar measures. At a national level, reducing the cost of
generating electricity by wind/solar power potentially bene…ts economic actors in other
countries and not just within the borders of the country where the innovation takes place.
Switching to low carbon activities requires the spread of low carbon technologies4
primarily in energy, infrastructure, transport and industry and e¤ective innovation in
these technologies in one nation or region will generate positive transnational externalities
(Human Development Report (2008)).
A case in point is the argument that the initial commitment to cut 50% of CFCs in
the Montreal protocol was critical to its success as it lowered the costs of making even
bigger reductions by providing a real incentive for the development of substitutes to CFCs
(Benedick (1998)). Thus, although there is a di¤erence in scale, the events surrounding
Montreal Protocol may be taken as a precedent for the potential of such a mechanism.
The point is unilateral commitments induce innovation, by creating a market for such
innovations, in technologies that could lower the cost of switching to low carbon economic
activities across di¤erent locations5 .
The positive spillovers generated by such unilateral initiatives, we argue, can be the
starting point for a new global policy initiative that enables a process of global learning,
both within and across countries, that delivers ever deepening emissions cuts over time.
We build a formal model where we analyze the conditions under which learning results
in global adoption of low carbon technology in …nite time. We show that while single
countries on their own may never get to the point of switching completely to low emission
activities, a suitably designed learning process evolving over strongly connected nations
will results in all countries switching to low emissions in …nite time.
Drawing on the results of the model, we discuss how policy design could a¤ect global
learning by impacting both the structure of interaction between countries and ensure
3
It is worth noting that even failure can generate useful information.
Technology is de…ned broadly here to include not only process and physical tech innovations but also
the associated behavioral/ cultural changes that go along with those innovations
5
The scope of such positive externalities may, however, be limited by issues of technology transfer and
absorptive capacity across locales in the face of binding political and cultural constraints.
4
3
integrated convergence to a global low emissions regime. Our results suggest that key
features a successful policy that builds on unilateral initiatives via global learning will have
to include developing a platform for exchange of information and subsidized monitoring,
strengthening spillovers across subnational actors in di¤erent countries and a new global
IP regime involving subsidized, targeted technology6 transfer of low carbon activities.
We critically examine a speci…c policy document presented by the G77 and China in
Copenhagen under paragraph 11 of the Copenhagen Accords which also addresses similar
issues. This proposal sets out a fast-track process for the di¤usion of relevant technologies
to either high emissions areas or those places were adaptation is already becoming a critical
concern. Building on our results, in section 4 below, we critically examine this proposal.
The remainder of the paper is structured as follows. Section 2 documents the variety
and extent of existing unilateral measures; section 3 sets out the model under which
global learning, building on existing unilateral measures, leads to a low emissions global
paradigm; section 4 discusses how aspects of this model could be usefully applied to the
post-Copenhagen process going forward, while the last section concludes.
2. How extensive are unilateral measures on emissions mitigation?
In this section we attempt to document, characterize and discuss the nature and extent of
unilateral measures which have been taken worldwide towards climate change. In tandem
with ongoing global negotiations, the major participant countries have also simultaneously
launched national action plans for combating climate change which involve extensive use
of unilateral commitments. To an extent, these unilateral commitments may represent
mechanisms for the implementation of proposed multilateral commitments, but in other
ways they are quite di¤erent. Thus, in the EU, there is a commitment to a 20-20 program,
which involves a 20% reduction in emissions and 20% of energy to come from renewables
by 2020. This is separate from multilateral commitments, though the EU has o¤ered to
go further to a 30% emissions reduction if other entities will match. Similar initiatives can
be found in the case of China, where there are extensive commitments to a 20% energy
consumption reduction relative to 2005 by 2020, a 45% reduction in carbon emissions
relative to GDP by 2020 and also a similar 20% commitment to renewable energy. These
forms of commitments, interestingly, also seem to involve deeper commitments by smaller
6
Technology is de…ned broadly here to include not only process and physical tech innovations but also
the associated behavioral/ cultural changes that go along with those innovations
4
countries. One striking case is Norway, which has committed itself to become a zero
carbon economy by 2050.
These unilateral examples are national but, on the face of it, go substantially beyond
what countries are seemingly jointly willing to commit to multilaterally; and beyond the
national level, there are also many further commitments also being made by sub-national
and local levels by governments, community based organizations, businesses, and even
by individuals. It is not unusual in Europe for individual cities to now have emissions
reduction targets by speci…ed dates. These could involve community based programs such
as free bicycle plans, as in Copenhagen, Denmark, or watershed renewal programmes, as in
rural communities bordering major rain forests as in Mexico, and community information
monitoring schemes as in sophisticated software which tracks carbon from individual
houses based on lifestyle and energy use (the idea being that increased knowledge will
change people’s behaviour). Commitments are also made at a business level, with new
businesses o¤ering what is needed for other businesses, communities and individuals to
"go green"7 . Because of the wider social/political commitment to emissions reduction,
it becomes good business to characterize products as emissions sensitive, contributing to
signi…cant emissions reductions. Finally, similar actions can be taken at an individual
level.
These unilateral actions being undertaken to combat climate change and reduce carbon
emissions around the world are both diverse and in constant ‡ux. The examples given in
this paper are representative samples of the levels and types of unilateral actions occurring.
2.1. Unilateral Measures by Country
In practice, national unilateral measures seemingly almost inevitably interact with multilateral negotiations. This is quite evident in the recent Copenhagen Accord document,
which relies quite heavily on unilateralism to achieve any sort of emissions reduction.
Many countries developed/ announced unilateral measures in the lead up to Copenhagen
and, while such commitments could be interpreted as a way of simply staking out bargaining positions, the argument we put forward is that, by including them in the Accord,
it increasingly causes the Accord to become a viable basis for an e¤ective global climate
change adaptation/ mitigation framework. A matter of particular interest, which we will
come back to in the next section, is the question of whether, by committing to speci…c
7
Recent estimates set the "green" industry worldwide at roughtly 360 billion USD, with
estimates that this could grow rapidly to 650-750 billion USD by late 2015.
See
http://english.cw.com.tw/article.do?action=shpw&id=10399&o¤set=0.
5
unilateral measures in the Accord, such nations positively alter the incentives of other
nations to commit to new or more stringent measures –in essence causing a snowballing
race to zero carbon on a wide scale. But, for now, we examine the breadth and scope of
such unilateralism that currently exists.
In addition to the examples of unilateral action already mentioned, a number of national governments and industries, particularly in Europe, are pushing for renewable energy in all its forms and renewables are gaining ground relative to other forms of power
production, albeit from a very small base. For instance, Denmark stands out as the wind
power capital of the world, especially notable since the push for the development of wind
power there predates the Kyoto Protocol. Similarly, New Zealand has developed a major
reliance on thermal power.
In some nations there is also a push towards increasing capacity in the generation of
nuclear power in addition to renewables. For example, in the UK, the national government, as part of its unilateral measures to cut emissions and achieve energy security, has
supported a new generation of nuclear power stations8 . Even though a nuclear path is a
somewhat controversial move for at least two reasons. First, there is no reliable method
for safely disposing of nuclear waste and second, given the large start-up costs involved,
investment in nuclear power may crowd out investment in other forms of green power
generation and use such as energy e¢ ciency and renewables. Another concern is that
current estimates of the supply of nuclear fuel predict only another 100 years before currently known global stocks are depleted. Thus, if low or zero carbon power generation
techniques do not become su¢ ciently commonplace in the intervening time, then nuclear
proliferation will only put o¤ the climate crisis, not solve it.
In the developing world, China has adopted a 45-20-20 plan to reduce emissions intensity per unit of GDP by 45%, as well as achieving 20% renewable energy use, both by
2020. A signi…cant step towards this is the phasing out of older ine¢ cient coal plants,
as well as an accompanying planned 20% improvement in energy e¢ ciency by 2020. India, as part of its National Action Plan on Climate Change, is actively seeking to alter
its energy sector to one that is ‘green’ via improved e¢ ciency, renewables (and nuclear
power), attempting to forge a ‘green’ development path which will take environmental
considerations into account with spillovers on other industries.
Similarly in the US, the focus is on the Waxman-Markey Bill, which sets out a 17%
emissions reduction relative to 2005 by 2020 and an 83% one by 2050 through an emissions
trading system and numerous related R&D, public awareness and institution creating
8
http://news.bbc.co.uk/1/hi/uk_politics/7179579.stm
6
projects. More recently, the Senate version of this, the Kerry-Boxer Bill, sets out goals
of a 20% and 83% reduction respectively for these dates, although it enjoys much less
support.
Sweden, Finland, the Netherlands and Norway have all imposed carbon taxes which
have been in place since the 1990s. In South Africa, enforcement of existing environmental
law has been an issue and hundreds of new environmental enforcement agents have been
trained to ensure prescribed environmental measures are implemented. In Brazil, there
is a plan dubbed the PROINFA plan, which aims to provide 10% of Brazil’s production
and consumption of energy by 2022 through renewables. Also there is Ecuador’s 2007
Yatsuni-ITT proposal, which seeks to preserve rainforest despite the oil below and to
slowly wean Ecuador o¤ oil dependence is indicative of the direction their government is
taking. These actions seemingly directly bene…t individual countries little relative to the
e¤ort involved.
Many of these national level unilateral actions, while reducing carbon, also serve more
local climate needs (adaptation-based projects), adding a self-interest based element to
proposed actions. For instance, in Spain, as part of a push to halt deserti…cation there,
a government led plan to plant 45 million trees over 4 years is underway9 . This serves
to reduce carbon emissions and lessen the country’s Kyoto shortfall but the need to
halt deserti…cation is the primary driver. A similar e¤ort, with salt-resistant plants and
de-salting plans as the weapons of choice, is underway against the more saline-based
deserti…cation progressing in Australia10 . As an example of what may occur globally
as extreme climate impacts occur, in the Maldives, the current president has set up a
sovereign wealth fund speci…cally to buy a new homeland, should the islands become
inundated11 via sea level rise saying, “We can do nothing to stop climate change on our
own and so we have to buy land elsewhere. It’s an insurance policy for the worst possible
outcome”. Thus, the phenomenon of unilateral actions in regards to CO2 reduction and
similar issues that cannot be explained by short-run self-interest is likely to be limited in
duration should climate change worsen and as more speci…c climate ‘damage’occurs.
This sampling of national unilateral initiatives is summarized in table 1.
(TABLE 1 HERE).
9
http://www.abc.net.au/news/stories/2008/09/13/2363739.htm
http://www.planetark.org/dailynewsstory.cfm/newsid/37246/story.htm
11
http://features.csmonitor.com/environment/2008/11/11/faced-with-rising-sea-levels-the-maldivesseek-new-homeland/
10
7
2.2. Unilateral Measures by Territories, States and Provinces
Unilateral measures are also evident at the subnational level. At the state and interstate levels of government in the US we see multi-state agreements such as the Mid-West
Greenhouse Gas Reduction accord, the Regional Greenhouse Gas Initiative, the Western
Climate Initiative and the Western Governors’Association –Clean and Diversi…ed Energy
Initiative. On the face of it, these initiatives are reactive to the US rejection of the Kyoto
Protocol. Setting restrictions on CO2 emissions by speci…ed dates is the content of all of
these save the Clean and Diversi…ed Energy Initiative, which sets a 30000 MW production
goal by 2015 for renewable energy among member states as well as more long term goals.
Many of these agreements are not con…ned to the US. Several Canadian provinces have also
signed onto some of these agreements, Ontario and Alberta for the most part. Within
Canada as well, there is a cap and trade plan being negotiated between Ontario and
Quebec, two of the top three emitting provinces.
Individual state e¤orts in North America have the common thread that nearly all
states and provinces have programs designed to improve energy e¢ ciency, although direct
emissions reduction e¤orts are the most common activity. With a large amount of carbon emissions originating in the power generation sector and North America having the
most energy intensive economy (per capita) globally, these two activities seem to be both
technical and political ‘low-hanging fruits’. The US states of California and Florida are
notable for e¤orts to improve energy e¢ ciency and CO2 emissions reduction in numerous areas including all of industry, transit/ autos and households, as well as opening the
market and experimenting with all types of renewable energies. Further north in Canada,
British Columbia is notable as the …rst province to independently impose a carbon tax
on itself.
In Australia, New South Wales has a $63 million (AUD) plan to overhaul the homes
of 220 000 low income families to improve energy and water e¢ ciency as a response to the
likely increased energy costs associated with Australia’s commitments under the Kyoto
Protocol. Most other Australian provinces are also improving energy e¢ ciency and some,
such as Queensland, are also actively and independently funding research into renewable
energy. Queensland’s program has already provided over $7 million (AUD) to numerous
projects on renewable energy production and e¢ ciency improvements.
Even in China and India, where provinces, by and large, walk in step with the mandates
passed down from the federal level, some regional initiatives already exists (such as solar
power and bio fuel projects in Karnataka12 ). One striking case is within Brazil where,
12
http://www.thehindu.com/2009/06/05/stories/2009060559100500.htm
8
somewhat in contrast to the national policies for increasing ethanol and biofuel production,
some states, particularly Amazonas, have shifted to rainforest preservation, going from
handing out chainsaws for free to ‘promote development’to now adopting the motto that
‘the forests are worth more standing than cut down’13 . In Russia on the other hand,
perhaps due to the fact that it trivially met its Kyoto targets, provincial unilateralism is
muted and mostly limited to data and information gathering14 .
Tables 2 and 3 summarize the unilateral initiatives discussed above.
(TABLE 2 here.)
(TABLE 3 here.)
2.3. Unilateral Measures by Cities
City level emissions mitigation e¤orts tend to lend equal weight to adaptation and mitigation, usually blending the two in proposed plans. In New York City for example, a
major initiative is underway to improve energy e¢ ciency and reduce emissions of all sorts
including a 30% reduction in greenhouse gas emissions. The focus is on replacing older
infrastructure with new and more energy e¢ cient technology which will also prepare the
way for any water, food, or natural disasters to be dealt with, and also to replace existing
cars with more fuel e¢ cient ones and to increase the number of trees and parks within
the city. Toronto, Canada has also engaged in energy e¢ ciency upgrading for many of
its buildings and infrastructure projects. Otherwise the plan of action is very di¤erent,
with Toronto seeming to prefer development of local sources of renewables over the larger
infrastructure upgrading projects and promoting green roofs heavily. Munich in Germany
has a plan very similar to Toronto’s, with the added …nancial innovation of weather derivatives regarding the weather’s favorability for generating renewable power in order to help
manage the risks involved15 . Similarly, London, UK, has a plan that focuses most heavily
on energy e¢ ciency upgrading, with projects on renewables.
In China, Shanghai has invested 80 billion Yuan (11.6 billion USD) in environmental
protection projects. The city, which is near to sea level has increased plant and tree
coverage to help ward o¤ erosion as the threat of ‡oods increases and is also intensely
focused in upgrading and installing infrastructure to ensure the city’s water supply. The
plan also provides incentives for promoting green industries within and around the city
13
http://www.sds.am.gov.br/dsv/download/img_download/20071226155008Relatorio%202.pdf
http://earth.esa.int/workshops/envisatsymposium/proceedings/posters/3P9/463731fr.pdf
15
http://www.munichre.com/en/ts/innovation_and_insurance_trends/
windmills_against_climate_change/default.aspx
14
9
and also has a goal for decreasing the volume of vehicles on the roads by 65%.16 Sydney,
Australia is focusing on water infrastructure upgrading, but other types of infrastructure
are secondary to the drive towards making the city carbon neutral via energy e¢ ciency
and renewable energy policies and related projects by 202017 .
A common thread in these unilateral emissions reduction initiatives (direct and indirect) is seen here in the form of a focus on renewables, energy e¢ ciency upgrading and
infrastructure renovation at a city level within the global sphere. On the other hand, more
so than at high levels of government, a clear focus on adaptive measures is also interwoven
into these policies.
Table 4 summarizes the various city-level unilateral initiatives discussed here.
(TABLE 4 here).
2.4. Unilateral Measures by Businesses and Joint Public-Private Partnerships
Unilateral measures are also being implemented by businesses and joint public- private
partnerships. While e¤orts similar to those mentioned above are ongoing in Mumbai (India) and in Moscow (Russia), unlike in most large cities around the world, most of the
e¤orts in these cities are often better described as business ventures rather than government e¤orts. This is because the deployment of Clean Development Mechanism (CDM)
related projects in these cities has tended to overshadow the need for explicit climate
change legislation in favour of legislation to manage a proliferation of CDM projects.
These often take a form similar to other city e¤orts, focusing on renewables, energy ef…ciency upgrading and infrastructure renovation, but with CDM projects subsidized due
to the carbon credits generated. These are not managed fully within the public sphere,
nor are they speci…cally limited to one city’s boundaries18 . This makes CDM a common
mechanism for the emergence of private-public partnerships in those areas where CDM
projects are valid.
One such project, for example, is the eco-friendly train project in Mumbai, where the
aim is that long term infrastructure viability will be maintained and government expenses
will be reduced by partnering with business to reduce train energy usage by over 30%.
16
http://en.chinagate.cn/development/environment/2008-12/15/content_16950071.htm
http://www.sydneywater.com.au/EnsuringTheFuture/ClimateChangeStrategy/Ensuring_the_Futire__Climate_Change_Strategy_-_Feb_2008.pdf#Page=1
18
http://www.articlearchives.com/energy-utilities/utilities-industry-electric-powerity/1816583-1.html
and
http://www.14000.ru/projects/city-climate/lea‡et0908e.pdf and http://bangalorebuzz.blogspot.com/2006/12/emission
cuts-to-augur-well-for-bmtc.html
17
10
Such projects take the place of pure public sphere e¤orts with the objective of greater
cost e¢ ciency for government and higher pro…t for business from the government. This
does, however, come with a greater degree of risk in terms of a project’s success or failure,
particularly when the project relies on unproven technological innovations that re‡ect the
desire to mitigate carbon emissions.
In some cases, CDM funding builds on existing locally owned private sector initiatives
such as the previously mentioned Karnataka Renewable Energy Project19 . While such
projects re‡ect an incentive to further local development goals (regardless of whether
those goals are intrinsically eco-friendly or not), the underlying incentives (driven by
CDM) still shape the projects so that they are ‘greener’ than what would likely occur
otherwise. However, that the requirement of “additionality” in CDM funded projects is
hard to verify in practice20 .
There are also large scale public-private partnerships, where the …nancial input of
CDM and similar incentive mechanisms is either less signi…cant or does not apply. In the
current global economy, when large businesses partner with governments at a national
or regional level, funding for large-scale experimental projects becomes possible. In the
United Arab Emirates, construction is underway for what is claimed will be the world’s
…rst zero carbon city (while in operation; construction is still just as carbon intensive).
This is notable for several reasons. First and foremost, the UAE is a country largely
dependant on production and sale of fossil fuels (mostly oil) and this project represents a
change from the status quo. Second, the scale of investment, which is roughly around $22
billion USD, makes it one of the most expensive green projects in the world. Third and
…nally, a key priority is given to pro…tability, in the hopes that this will signal to others
that similar projects can be pro…table in other areas of the world as well21 . The embryonic
city, which lies near Abu Dhabi, has infrastructure designed to encourage energy saving
and estimates are that it will use 75% less power than a conventional city of the same
size, and all energy that is used will be generated using renewables22 .
19
Ravikiran Power Projects Private Ltd,
Monitoring Report:
7.5 MW GridConnected Biomass Power Project (UNFCCC Ref.
No.
0971), Version:
Ravikiran / 001, dated 4 September 2008 (‘Monitoring Report’), p.
8, available at:
http://cdm.unfccc.int/UserManagement/FileStorage/13KBH5NLQM60FOWXY9PZ8J47TCISAD,
accessed on 31 March 2009
20
We discuss the shortcomings of CDM as a funding mechanism for technogy transfer in greater detail
in Section 5.1.
21
This is an example of the demonstration e¤ect, a point that we will come back to later on in the
following section.
22
http://www.dailygalaxy.com/my_weblog/2008/05/the-worlds-…rs.html
11
Business involvement in emissions mitigation also partly re‡ects the perceived profitability of going ‘green’. For example, in Dubai, a company called Dynamic Architecture23
is working on rede…ning skyscraper construction in terms of e¢ ciency and also in terms
of self-sustainability, suggesting new building methods as well as making skyscrapers into
wind power generating plants, able to power themselves and several city blocks at a fraction of the cost and construction time of a more traditional skyscraper. Another example
is Interface Carpets, a carpet manufacturer that boasts the so-called ‘greenest CEO in
the world’, Ray Anderson. Carpet making is oil and carbon intensive and Mr. Anderson
has managed to reduce carbon emissions in their process by 82% over the past 12 years
in absolute tonnage, while at the same time doubling pro…ts. Approximately, this is the
reduction goal for the whole world set out for the long term under the UNFCCC negotiations to keep warming at 2 degrees. Thus a case study of this company makes for a good
example of how to reach that goal. Perhaps even more signi…cant, it stands as a signal
that the stated international goal is reachable even for fossil fuel intensive industries –
and at a pro…t as well24 .
Similar businesses are growing on a global scale and a large subsection of them, mostly
from developed countries, attend the annual ‘Clean Equity’25 conference in Monaco, where
green industry leaders mingle to share ideas and network with each other. Here gather
companies developing greener cars, better batteries, better solar cells, consultancies o¤ering services on how to green your business, concept businesses for launching new geothermal, wind and solar energy ideas, companies selling green gimmicks26 and several other
types of green companies.
Perhaps somewhat counter-intuitively, oil companies have also embraced the statement
that going green is pro…table. This is embodied in the US Carbon Action Partnership
(USCAP)27 ; a large group of oil companies, mining companies and other heavy emitters
who are essentially calling for regulation to be placed over them in terms of carbon
emissions, calling for signi…cant and early action and for the US (home to all head o¢ ces
of the members) to take an early and signi…cant stance on reducing carbon emissions.
23
http://www.dynamicarchitecture.net/home.html
Mr.
Anderson’s speech on this at a recent TED conference may be found at
http://www.youtube.com/watch?v=iP9QF_lBOyA
25
http://www.cleanequitymonaco.com/
26
Things such as this idea from Honda, opening up more car options as your driving gets greener, would
qualify as a gimmick. http://earth2tech.com/2009/03/02/upcoming-honda-insight-turns-eco-friendlydriving-into-game/
27
http://www.us-cap.org/, some members include Shell, Chrysler, DuPont, Dow Chemical, GM, Ford
and GE
24
12
While the purpose here is to guide change so that the most pro…t/ least damage is won,
that such an agreement exists at all is indicative how important these companies believe
emissions reduction to be, whether this stems from ecological concerns or worries over
their bottom line pro…t.
2.5. Unilateral Measures by Individuals
Individual actions are also having a signi…cant impact on climate change mitigation. Recycling is increasingly becoming common across the globe. In many developing countries,
recycling is a source of livelihood for the urban poor28 . Compact ‡orescent bulbs are
also becoming more commonplace every day, with LED technology, perhaps even more
e¢ cient, progressing apace. Ultimately, the goal is to make households as energy e¢ cient
and carbon free as possible. For the dedicated, this may involve purchasing for energy
e¢ ciency, while installing home-based solar or wind power systems. If individuals have
the will, funds and supporting legislature this set of behavior is becoming a global commonality. Individuals also in‡uence higher levels of governance, businesses and whole
groups of other individuals. In simpler terms, these individuals attempt to shift culture
more towards the ‘carbon free’ mentality. Some who have put notable e¤ort into this
include signi…cant political …gures such as Al Gore and Arnold Schwarzenegger of the US,
Nicholas Stern of the UK and Chancellor Angela Merkel of Germany. Some celebrities
and other well known people, with varying degrees of e¤ectiveness, have also advocated
climate change action and left a mark, albeit a smaller one. These include David Suzuki,
Sir Paul McCartney, Jimmy Bu¤et, Celine Dion, Keanu Reeves, Alanis Morrissette and
countless others, a large number of them well known internationally.
Of other individuals, completely unknown before the climate change issue arose, there
are many who have had impact. Perhaps one of the most e¤ective is Greg Craven. Using
Youtube as his initial medium, this small town physics & chemistry teacher has created,
…rst an ominous initial statement on climate change entitled “The most terrifying video
you’ll ever see” which has since generated several million hits, followed by a carefully
planned and thought out 7 hour long series of videos entitled “How it all Ends”that expands on the initial message, responds to critiques, and discusses and explains the science
climate change for the average person in a way that cuts through all the media sensationalism and confusion. His argument for action on climate change was so compelling
that several international news stations from around the world became interested and
he has also seemingly been contacted by a lead author of the Stern Review. His stated
28
http://www.guardian.co.uk/environment/2007/mar/04/india.recycling
13
intent is to make his message ‘go viral’via as many media formats as possible and change
mainstream culture in regards to views on emissions and climate change in general. These
examples illustrate the point that individuals can, and do, in‡uence large groups of people via existing social networks and, through them, seek to alter the mainstream culture
enough to in‡uence higher levels of authority and government.
3. Unilateral Measures and Global Learning
In this section, we examine the conditions under which global learning, building on the
positive spillovers generated by unilateral actions, delivers ever deepening emission cuts
over time and convergence to a low carbon world in …nite time.
Rationale for unilateral measures:
A number of the examples discussed in the preceding section make the point that it
could actually be in the self-interest29 of relevant economic actors to undertake unilateral
measures that also result lower emissions. Nevertheless, the diversity of unilateral initiatives re‡ects a underlying heterogeneity of interests, beliefs, motivation at the level of
countries, regions and groups.
Certain forms of collective identity can be self-enforcing in that conditional on other
individuals accepting the same collective identity, it is in the self-interest of any one
individual not to deviate, a point emphasized by Olson (1971) in his work on collective
action30 . A di¤erent possibility is that the agent who cuts emissions obtains a "warm
glow" as discussed in the literature on philanthropy (see, for example, Andreoni, 2006)
from the act of cutting emissions.
Straightforwardly, an economic agent will unilaterally cut emissions if the private
bene…t from "going green" is greater than the cost of cutting emissions. Importantly, the
cost of cutting emissions could limit, at any one point of time, the size of the group of
individuals, within a given population, who would be willing to unilaterally undertake
costly emissions cuts. Although a national level commitment to emission cuts may not
emerge as the outcome of majority voting, unilateral measures may then exist at various
29
Examples of which include adaptation to the local impacts of climate change, ensuring energy security,
halting the process of deserti…cation, local development needs.
30
A related rationalization of unilateral initiatives lies in "rule utilitarianism" (Harsanyi (1977)) where
individuals act to conform to a speci…c rule given that some group of other individuals also conform to
the relevant rule. Unilateral measures in a given group can be rationalized if each individual in that
group …nds its optimal to cut emissions given that all individuals in that group conform to the rule of
cutting emissions
14
subnational levels.
A formal model of unilateral measures, global learning and cumulative
emission cuts
Formally, we consider a model with discrete time intervals which go from t = 1; 2; :::.
There are n = 1; :::; N countries. Each country consists of a number of individual entities
P
of mass mn > 031 and the total mass across countries is M = i mi . At t, etn;j 2 f0; 1g
denotes the emissions of greenhouse gases by individual j belonging to country n so that
at any t; etn;j = 0 corresponds to adopting low carbon activities while etn;j = 1 corresponds
Rm
to persisting with a high carbon activities. Let etn = j i etn;j dj denote the total emissions
P
of country n in period t. Let E t = i etn denote the total emissions at time t.
t
Let a fraction tn (respectively, 1
n ) of individuals j in country n have preferences
an incentive to cut emissions unilaterally (respectively, continue with high emission activities). We assume that once a switch occurs from high to low emission activities it is
irreversible.
We assume that in each country n at each t, individuals decide, via majority voting,
between two alternatives: (i) all individuals within the country switch to zero emission
activities, or (ii) only those individuals who voluntarily choose to do so switch to low
emission activities. Therefore, a country n will commit to cut emissions to zero in period
1
t (and in all subsequent periods) i¤ tn
. Otherwise, the emissions level in country n
2
t
t
at t is given by en = (1
m )mn .
0
We will assume that n = n < 12 so that no country is willing to commit to emission
cuts as part of a multilateral agreement.
We model the preferences and behavior of individuals within a country as follows.
Assume that individual j in country n obtains a private bene…t bj 2 [0; B], B > 0 with
individuals within a country uniformly distributed on the interval [0; B]. Consistent with
our earlier discussion, we interpret this private bene…t to the individual from cutting
emissions as derived from either group membership or a "warm glow" or just the direct
P
t 1
bene…t from successful local adaptation. Let cn
denote the cost of cutting
n0 n;n0 n0
emissions to individual located in country n given that a proportion tn0 1 , n0 = 1; 2; :::; n
of individuals in all countries have already cut emissions at t 1 and n;n0 , 0
1,
n;n0
captures the spillover from country n0 to country n; with n;n = 1.
We assume that cn (:) is strictly decreasing function so that a unilateral commitment
31
We interpret mn as the size of country n. This is used as an approximate measure of a country’s
ability to in‡uence carbon levels. Formally, we assume that nation n consists of a set of individuals n of
Lebesgue measure mn .
15
by a group of individuals within country n at each t to cut emissions in any time period
stimulates innovative activity that lowers, over time, the relative cost32 of adopting low
carbon activities for other individuals both within country n and across all other countries
n0 6= n.
We justify this key assumption in our model as follows. By undertaking unilateral
measures economic actors demonstrate not only the feasibility of collective action and
their awareness of the potential threats to other actors but also lower the cost of cutting
emissions for all other actors as well. Thus Heal (1993) argues that as one country undertakes even limited emission cuts it incurs a variety of costs (e.g. R and D investments,
retooling) that are "sunk" in nature. However, once the new low carbon technology has
been developed, it can be made available to another country at a relatively lower cost.
Moreover, given the larger market, there are greater private incentives to innovate in both
countries leading to deeper emission cuts within the two countries and at some point, inducing emission cuts by a third country and so on. The general point is that as one
country cuts its emissions, the marginal cost of cutting emissions for other countries may
fall as well thus making emission cuts more worthwhile in these latter countries.
In the OECD, Europe and large population, rapidly growing developing countries over
50% of industrial emissions are accounted for by 3-4 industries (power generation foremost), and hence attempts to cut emissions and/or the adoption of technology standards
in these key sectors will have a signi…cant potential impact on overall emissions not only
within these countries but also in other countries.
The spillover e¤ects across countries, for any one country n, is captured by the term
P
t 1
n0 n;n0 n0 . Speci…cally, both the sign i.e. whether or not n;n0 > 0 and the magnitude
of n;n0 for a pair of countries n; n0 will be a consequence of the use of appropriate policy
mechanisms that build on positive technological spillover across countries. Such a policy
will be designed to maximize the impact of positive technological/institutional externalities to facilitate innovation, technology transfer and adoption of low emission activities
both within a country and across countries. Subsidized targeted technology transfer may
be essential to alter the participation constraints of nations over time. This will be the
focus of the discussion on policy in the following section.
P
t 1
Let gnt = j : bj cn
0 denote the group of individuals for whom
n0 n;n0 n0
it becomes a dominant action to cut emissions at time t within a nation n. De…ne the
32
These costs could be …nancial costs, lowering the relative costs in terms of the e¤ort or time sacri…ced
to do the ‘green’thing i.e. making it easier in terms of the physical and cognitive e¤ort involved to do
the ‘green’thing rather than the more carbon intensive equivalent action.
16
function
Gn
X
t 1
i;j j
j
!
(
=
j : bj
cn
X
t 1
n;n0 n0
n0
!
)!
0
where (:) is the Lebesgue measure. Thus, Gn (:) denotes the proportion of individuals
in country n for whom it becomes a dominant action to cut emissions at t.
We will assume that Gn (x) = 1 for all x
1. This is equivalent to requiring that
cn (x) = 0 for all x 1 a strong assumption that will simplify the statement of the main
result. In e¤ect we are assuming that all internal and external learning will cause an
increasing proportion of individuals within a country to cut emissions (or at least, any
negative in‡uence is always overridden by the positive in‡uences). After the statement and
proof of the main result below, we will show how our result is robust to this assumption.
Finally, we assume that underlying preferences (which we do not explicitly specify) are
such that it is globally e¢ cient to set etn = 0 for all n; t. The divergence between private
payo¤s and global payo¤s is assumed to re‡ect the wide-spread negative externalities (the
medium term damage caused by global warming) resulting from continued high emission
activities.
2, by the
The evolution of tn within each country n over time is described, for t
system of di¤erence equations :
t
n
= min
(
t 1
n
+
n
Gn
X
t 1
n;n0 n0
n0
!
t 1
n
!
)
; 1 , n = 1; :::; N , (1)
where n , 0 < n
1, is a measure of inertia in learning within country n: not all
individuals within country n who bene…t from switching actually do so- this could be
due to variety of factors such as habit formation, lack of awareness and information etc.
Note that (1) is a system of di¤erence equations that cannot, in general, be reduced to a
collection of uncoupled di¤erence equations, one for each country n.
We assume that the map G = fGn : n = 1; :::; N g, G : [0; N ]N ! [0; N ]N , is continuous
and increasing on [0; 1]N : learning within and across countries is complementary.
Will each country eventually be in a position to commit to cut emissions? We say
that country n is connected to country n0 if n;n0 > 0. De…ne a directed graph over N
where the vertices are individuals and an arc (n; n0 ) exists i¤ n is connected to n0 . A path
in a directed graph is an ordered collection of arcs and vertices in which all vertices are
distinct. A directed graph is strongly connected if for every pair of distinct if for every
pair of distinct vertices (n; n0 ) there exists a path connecting n to n0 . We say countries
are globally strongly connected if the associated graph is strongly connected.
17
An example of a scenario where countries are globally strongly connected is when
0
n;n0 > 0 for all n 6= n (the complete graph) so that no one country is pivotal in driving
the global learning process. A di¤erent strongly connected scenario is one where n;1 > 0,
0
0
1;n > 0 but n;n0 = 0 for n 6= n whenever n 6= 1 (the star graph). In the latter case,
country 1 (like the US) is pivotal in the global learning process as any connection between
two other countries must go through country 1. Other possibilities include scenarios where
a group of countries (e.g. US, countries within the EU, the BRICs) are pivotal in ensuring
that the countries are globally strongly connected.
More generally, when countries are globally strongly connected, between any two countries n; n0 , there is a chain of countries n0 = n; :::; nS = n0 with ns ;ns 1 > 0, s = 1; :::; S.
The following proposition states the conditions under which global learning, building
on unilateral actions within some country, delivers a switch to low emissions in …nite time:
Global Learning Result: Suppose learning within each country is described by the
map (1), countries are globally strongly connected and n > 0 for some country n. Then,
there exists t^n , decreasing in n , n;n0 , n; n0 = 1; :::; N , such that country n will commit
t
^
to cut emissions for all t
t^n , with etn = (1
n )mn for all t < tn . A stable global
agreement to cut emissions will emerge at t^ = maxi t^n .
Proof. Suppose the set of countries is globally strongly connected with n > 0 for
some country n. Consider the map G = fGn : n = 1; :::; N g where G : [0; n]N ! [0; n]N .
Note that by construction Gn (x) = 1 for all x 1, n = 1; :::; N . Therefore, the vector
(1; :::; 1) is always a …xed-point of G.
P
0
Note that Gn
n0 n;n0 n > n whenever 0 < n0 < 1 for some n 6= n with n;n0 > 0.
Whenever tn0 > 0, for some country n0 at some t, as there is a path connecting each
P
t0 1
country n to country n0 , n0 6= n, there exists t0
t such that Gn
>
n00 nn00 n00
t0 1
n
0: t = t0 if tn > 0 and nn0 > 0, otherwise t0 > t, t0
t + s where s is the
minimum number of vertices (the minimum taken across all the paths connecting n to n0 )
between n and n0 .
As n > 0 for some country n, it follows that (1; :::; 1) is the only …xed-point of G.
Let f t : t 1g denote the sequence generated by an iterated application of the RHS of
(1). Then, by continuity of G, for each country n supt tn = 1 and as f t : t 1g is a
component-wise increasing sequence, limt!1 tn = 1 for all n = 1; :::; N .
1
Let t^n = inf t : tn 12 . Therefore, for all t
t^n , tn^n
and a stable global
2
^
^
agreement to cut emissions will emerge at t = maxn tn . As the RHS of (1) is increasing
in n , n;n0 , n; n0 2 N , for each country n, t^n is decreasing in n , n;n0 n; n0 2 N .
The above result implies that as long as there are some (however weak) spillover
18
e¤ects in learning across countries n so that all countries are globally strongly connected,
eventually a majority of voters in each country n will vote to commit to cut emissions.
Some remarks:
1. In the proof of the preceding proposition, we assumed that all learning within
and across countries is complementary so that G is increasing on [0; 1]N : all internal and
external learning will cause and increasing proportion of individuals within a country
to cut emissions (or at least, any negative in‡uence will tend to be overridden by the
positive in‡uences). It is straightforward to note that nothing essential in the preceding
proof requires the G to be increasing over the whole of [0; 1]. What is essential for the
argument is that any …xed-point of G is greater than (in the usual vector ordering) the
N
so that negative in‡uences
vector 12 ; :::; 12 which requires that G is increasing on 0; 12
dominate positive in‡uences when a su¢ ciently large fraction of individuals within each
country have already switched to cutting emissions.
2. While the formal model suggests that with spillovers all countries will eventually
agree to cut emissions down to zero, there still remains the issue of how quickly countries
P
1
get there. For example, suppose n = 1 and n n
: if nn0 = 1 for all countries
2
1
0
t
n; n , then n
for all n for each t
2. This is clearly a best-case scenario, one
2
where convergence is immediate: t^ = t^n = 2. More generally, as for each country n, t^n
is decreasing in n , n;n0 , n; n0 = 1; :::; N , policy interventions that increase the values
of these variables (i.e. reduce learning inertia within a country and strengthen spillover
e¤ects across countries) will increase the rate at which there is global convergence to a
low emissions regime, and even if these cannot be in‡uenced on a national level due to
economic or political realities, it may still be possible to alter policy at a subnational level
in some instances.
3. Implicit in our formulation is the assumption that individuals within a country
behave myopically. If, on the other hand, individuals in a country anticipate that unilateral actions lower the cost of switching to low emission activities for individuals in other
countries, even if it isn’t in the direct short-term interests of certain actors to cut emissions, by anticipating that such activities will generate a similar response from others,
such actors may undertake unilateral measures33 thus speeding up the process of convergence to a global low carbon regime. Economic actors in countries that (a) create the
largest spillover e¤ects, either directly or indirectly, on global learning, (b) are pivotal
33
A simillar result has been obtained in a model of farsighted network formation in Dutta, Ghosal and
Ray (2005). Chatterji and Ghosal (2009) also make a simillar assumption in the context of a discussion
on climate change.
19
(i.e. without whom global learning will be delayed substantially), and (c) are willing to
bear the costs of being one of the …rst to switch to low emission activities, are more likely
to act in anticipation of inducing an earlier switch to low emission activities by other
countries. Moreover, such behavior will be in‡uenced by the strength of the spillover in
learning across countries. For example, if n = 1, nn0 = 1 for all countries n; n0 , voters in
1
country n could choose to cut emissions to zero (so that 1n = 1) anticipating that tn0
2
0
for all n 6= n, t 2.
To understand the importance of spillover e¤ects in driving global learning to the point
where countries commit to switching to low emissions, it is useful to consider the polar
opposite case, where nn0 = 0 for all n 6= n0 : in this there are no cross-country spillovers
and each country is isolated. In this case, all learning takes place within a single country
and the evolution of tn within each country n over time is described, for t
2, by the
equation:
t
n
= min
t 1
n
+
n
Fn
t 1
n
t 1
n
;1
n = 1; :::; N
(2)
where Fn (:) = Gn (:) on [0; 1]. Clearly, in this case, (2) is a system of di¤erence equations
that can be reduced to a collection of uncoupled di¤erence equations, one for each country
n each of which can be analyzed separately.
In this case, will each country eventually be in a position to commit to cut emissions?
The answer in general is no. Let n = min f 2 [0; 1] : Fn ( ) = g is smallest …xed point
of the map Fn : [0; 1] ! [0; 1]. n is well-de…ned as 1 is always a …xed point of Fn (:) and
the set of …xed-points of Fn (:) is closed subset of a compact set and hence, compact. Let
f tn ; t 1g denote the sequence generated by an iterated application of the RHS of (2)
with 0n = n 0 and n
n > 0. Clearly, if n = 0, country n never moves away from
0. If n > 0, then by continuity of Fn (:), supt tn = n and as f tn ; t 1g is an increasing
sequence, limt!1 tn = n . Therefore, as long as n < 21 learning within country n will
never result in country n, as a whole, switching to low emissions.
More generally, it is straightforward to construct examples where limt!1 tn < 12 for
most countries when countries aren’t globally connected. Suppose 0 < 1 < 12 , n =
n1 = 0, n 6= 1. Then, no country other than country 1 moves away from 0 as country
1 is isolated (i.e. not linked to any other country) and this is the only country where
unilateral actions are being undertaken. Clearly, other less extreme examples, along the
above lines, can be constructed.
In conclusion, our formal analysis has shown that while single countries on their own
may never get to the point of switching completely to low emission activities, a suitably
20
designed learning process with strongly connected nations will, over time, deliver a global
switch to low emissions.
In the following section, we will use these results to develop a policy proposal that
explicitly accounts for the possibility that global learning, by building on unilateral measures, could, in principle, deliver a switch to a low carbon world.
4. Global learning, the role of technology and IP, and the road
to a low carbon paradigm
In this section, we use the results of the model developed in the preceding section to
discuss key features of a global policy regime that ensures cross-country convergence to a
low emissions regime by e¤ectively building on the positive externalities inherent in the
unilateral initiatives.
In practice, national unilateral measures almost inevitably interact with multilateral
negotiations. This is quite evident in the recent Copenhagen Accord document, which
relies quite heavily on unilateralism to achieve any sort of emissions reduction. Many
countries announced unilateral measures in the lead up to Copenhagen and, while such
commitments could be interpreted as a way of simply staking out bargaining positions,
the argument we put forward is that, by including them in the Accord, it increasingly
causes the Accord to become a viable basis for an e¤ective global climate change adaptation/mitigation framework. By committing to speci…c unilateral measures in the Accord,
how can such nations positively alter the incentives of other nations to commit to new
measures –in essence causing a snowballing race to zero carbon on a wide scale?
To begin with, there is the issue of who the participants in global negotiations should
be. Given that subnational groups are more likely to have the autonomy to commit
resources in initiating unilateral measures, an open question, therefore, is whether subnational groups, such as provinces, states, or territories, who can exercise such autonomy
in otherwise non-participating nations could be allowed to include their commitments
directly in a global agreement post-Copenhagen.
Central to the model and the Accords is the idea of a global learning process, in
which technology and innovation …gure prominently. An alteration to the way technology
transfer works on a global scale has already been proposed under paragraph 11 of the
Copenhagen Accords. This proposal, which was presented by the G77 and China in
Copenhagen, sets out a fast-track process for the di¤usion of relevant technologies to
either high emissions areas or those places were adaptation is already becoming a critical
21
concern. As set out currently, this to be governed by an Executive Body on Technology
which will operate under the authority of the COP and operate using a new fund called
the Multilateral Climate Technology Fund, largely …nanced by Annex II countries but
supplemented by Annex I contributions, with the incentive being that contributions to the
Fund would count towards a country’s Bali Roadmap responsibilities34 . The proposal also
sets out to accelerate the rate at which research and development on such technologies is
conducted and to …nance it through venture capital and aid in rapid commercialization and
di¤usion whereupon, presumably, a portion of the funds devoted to the innovation would
be recovered back into the multilateral pool. There is an inherent selection bias in this,
since the Executive Body is selecting the innovations to go forward to commercialization,
but this might be possibly minimized by the makeup of this group and the oversight of
COP.
In what follows, we examine key elements of this proposal in greater detail.
While the basic framework of this proposal is already present, there is little discussion
so far of veri…ability aside from the fact that the Executive Body would examine each
case. A key issue that may arise is one that previously arose with the Clean Development Mechanism (CDM) under the Kyoto Protocol, that is the issue of how to ensure
"additionality". In practice, this has been hard to verify. For example, …rms would delay
adoption of cost-e¤ective low carbon technologies to bene…t from CDM or use CDM to
adopt technologies that they would have funded from capital markets or internal funds in
any case(Olsen (2007), Wara and Victor (2008)). And so, missing in this proposal thus
far is the issue of what sort of conditions should be attached to these payments to ensure
"additionality". In the case of carbon mitigation technologies these conditions could be
in the same terms as are commonly being made by nation states. That is, in terms of
time bound carbon emission or carbon intensity targets, and this could be particularly
useful in key sectors such as energy, infrastructure, transport and heavy industry.
Also, a conspicuously missing element of the proposal is how, exactly, it would interact
with the currently existing intellectual property rights regime. What could be involved
is a new global IP regime characterized by governments in countries developing publicly
funded new technology that involves users in other countries at the development stage,
paying a part (or whole) of the royalties paid by users in other countries for privately
funded technology, joint ventures between relevant actors (either in the private and/or
public sector) across national boundaries and lowering the cost of local use (and adap34
A key advantage of this is that the amount of funding provided is not dependent on the price of
carbon, allowing the ‡ow of funds to be much more stable than otherwise.
22
tation) of proprietary technologies by designing an appropriate international licensing
system.
In order to qualify for funding in the proposed global IP regime countries would have
to adopt speci…c commitments i.e. speci…c time-bound quantity targets like initially
lowering carbon intensity followed by emission cuts, the adoption of low carbon technology
(carbon capture and storage, solar and wind energy etc.) in important sectors such as
energy, infrastructure, transport and industry. The funding of individual projects within a
qualifying country could decentralized so that royalties paid by users for privately funded
technology originating elsewhere are refunded and the cost of local use (and adaptation)
of such proprietary technologies subsidized. A portion of the funding could be reserved
for research and development. Decisions on funding individual projects would be taken
in the context of an overall carbon reduction plan for the country as whole. Institutional
design to ensure timely funding ‡ows, monitoring and evaluation would be essential to
the successful operation of such a global innovation and technology transfer fund.
The details of this relation will be critical in determining the e¤ectiveness of this
proposal as, at present, innovations of the type needed particularly in the key sectors
mentioned above typically have to go through years, and sometimes decades, of testing
and red tape before they can become commercially available. If this generally remains the
case under the proposed Technology Mechanism, then its e¤ectiveness towards mitigating
carbon and other emissions in a timely manner will be questionable, and to a lesser extent,
this is also true of technologies related more towards adaptation. This process of testing
and red tape, along with issues of adaptive capacity in countries that technologies di¤use
to, is one of the main issues that raised the need for a fast-track di¤usion process in the
…rst place.
While not included in the proposal as it stands, the strength of the links (positive
spillovers) between countries in this process would be very important in almost every
aspect. For example, the US and the EU would be especially important because of their
central role in the world economy and their generation of innovation and technology
transfer. Others such as China and India would be important because of the size of their
populations and potential for emissions mitigation. For example, existing "clean coal"
power plants and carbon capture technologies can be developed and further re…ned in the
US and EU with a subsequent transfer to China under the mechanism proposed where
it would have a signi…cant impact on cutting emissions. Other links may re‡ect structural similarities, land use patterns, existing patterns of carbon consumption, etc. Such
a structure, if included in the current proposal, would aid in the e¢ cient distribution of
funds by allowing them to be distributed according to characteristics such as the degree
23
of spillover by type of technology and by targeting countries that have the greatest potential to generate spillover e¤ects35 . Such a policy regime would involve platforms where
information relating to unilateral initiatives can be exchanged and is likely to involve
subsidized monitoring.
Identifying, quantifying and calculating such spillovers will be key to the success of
the proposed mechanism and could feasibly …t into the mandate of the Strategic Planning
Committee within the proposed Executive Body on Technology. This will be essential,
as encouraging the type of technologies needed in generating positive spillovers on these
across countries are both likely to be costly and trade-o¤s may become necessary.
In general, measures that reduce emissions inertia within a country and measures that
strengthen the positive spillover e¤ects across countries are both likely to be costly. With
resource constraints, there is likely to be a trade-o¤ between the two such as subsidizing
measures that improve the energy e¢ ciency of domestic households or subsidizing low
carbon technologies that reduce emissions in the energy generation. The latter may have
a higher potential to generate spillovers across countries while the former will have a
bigger impact on reducing inertia within a country.
A further minor complication, as stated previously, is the fact that the COP has not
o¢ cially adopted this proposal. The phraseology within it assumes that the Mechanism
will answer to the authority of the COP, and as stated in the introduction, the way things
played out in Copenhagen suggests that it may be possible to move this process forward
more e¤ectively and e¢ ciently through a body other than the COP and the UNFCCC
such as the G20 or even just the US and the BASICs. Depending on the way things play
out the structure of the proposal may have to be changed to re‡ect this.
But despite these complications the proposal has the potential, if the spillovers we
mention can be created in su¢ cient strength, to speed the convergence to a global low
carbon regime signi…cantly, particularly if the largest global actors in the climate change
issue indicate a willingness to become involved.
The process described here may take time to play out so that the emission cuts required to stabilize global temperatures may not be delivered quickly. We envisage a
process of technology di¤usion that involves chains of innovations with new inventions
based on other low carbon technologies: such a process may require roughly 5-10 years
to play out for each innovation, or more if the innovations become controversial in some
35
This would be due to speci…c characteristics such as size, in‡uence, technological and innovation
capabilities, the degree of similarity with other national economies such as location, patterns of land and
energy use, dominance of key sectors, neighborhood e¤ects etc.
24
way. Some of these new technologies will not be compatible with high carbon technologies
and entire factories may need to be retooled (thus raising the adoption costs of the new
technologies). Although innovation and subsequent transfer of new technologies is essential, emissions reduction may be achieved by ensuring the spread and adoption of existing
low carbon technologies within and across countries. For example, households and …rms
within high income countries could be persuaded to insulate their houses or install solar
panels by a combination of subsidies (or low cost loans), and by the extension of carbon
markets to individual households and small …rms/businesses. As already argued, subsidized technology transfer across countries (for example, the transfer of existing cleaner
and more e¢ cient power generation from coal to China and/or the greater use of carbon
capture technologies in power generation globally) would require a new global IP regime.
If cross-country spillovers occur on the level this proposal would need in order to be
considered a success, and almost inevitable question arises of what this would do to trade
‡ows. If this Mechanism is to be a central feature of global climate change adaptation
and mitigation e¤orts the implication is for a further rise in globalization and, to a limited
extent, some integration between the trade and climate change spheres. This would have
a number of bene…ts, not the least of which is giving the private sector incentive to be
active on the climate change issue through the proliferation of innovations necessary for
a global low carbon paradigm.
It also suggests an eventual enforcement mechanism in the form of trade sanctions or
increased protectionism (or even the threat of such) against free riders or those who refuse
to participate. Bundling trade and climate change negotiations together could ensure
broader participation and compliance in climate change negotiations because the ‡ow of
immediate bene…ts associated with emission cuts (the bene…ts of lower trade barriers)
could alter incentives for countries to participate in global negotiations. However, a
necessary condition for such bundling to work is that the threat of increased protectionism
by low carbon nations be renegotiation proof, a condition (i.e. increased protectionist
tendencies) more likely to met by nations already undertaking unilateral initiatives as
demonstrated by recent talks on carbon based border adjustments to address issues of
leakage.
Finally, note that to achieve the goal of ensuring that the pace of innovation in low
carbon technologies is rapid, such a policy would have to be supplemented by other
measures. If there is uncertainty over commitments to emission targets and carbon prices
‡uctuate over time, or are too low or if too many economic activities are excluded from
emissions trading, there may be little or no impact on behavior of …rms and households.
This may discourage innovation (costly investment in the production of new ideas) that
25
lowers the relative cost of low carbon activities in the …rst place.
5. Concluding remarks
In this paper we discuss both the size and extent of the unilateral commitments to reduce
carbon emissions which are being taken in countries around the world and their potential
for inclusion in the emerging Copenhagen Accord process. The advantage to this over
previous attempts at a multilateral climate change agreement is, of course, that if multilateral commitments are to be largely met through unilateral processes (particularly ones
already underway or planned), then commitments taken on by negotiators involve little
additional cost. But more broadly the question is whether unilateral measures can act
as an e¤ective engine for reduction in carbon emissions and achieve the goal of limiting
temperature rise to 2 C. Ultimately, this will depend on unilateral measures’relative effectiveness, their number, and their ability to create the synergies necessary to reduce the
cost for other economies globally to follow suit and switch to low carbon via process of
global learning. For such a process to be successful, the post- Copenhagen process going
forward should attempt to build on both national and subnational e¤orts to promote the
development and spread of e¤ective unilateral action and encourage implementation of
policy that strengthens the spillovers of such actions across nations, which is key to the
social learning process we have previously described in the model. Absolutely key for this
process will be changes to the e¤ective international intellectual property rights regime of
the sort described in paragraph 11 of the Accord. A less restrictive regime is critical for
allowing spillovers across nations to take place on a timescale, according to the current
science, that would be meaningful for dealing with climate change. Thus contrary to
the popular view of the Copenhagen Accord being empty and a papered over agreement,
in the context of the model described here, the Accord has the potential to become a
solid foundation for an e¤ective post-Kyoto global climate change regime. Whether that
potential will be realized remains to be seen.
References
[1] Andreoni, J, 2006, “Philanthropy” in L Gerard-Varet, S Kolm and J Ythier (eds.),
Handbook for Giving, Reciprocity and Altruism, Elsevier: Amsterdam.
[2] Barrett, S. (1994), "Self-Enforcing International Environmental Agreements", Oxford
Economic Papers, 878-894.
26
[3] Chatterji, S. and S. Ghosal (2009), "Technology, unilateral commitments and cumulative emissions reduction", forthcoming CESIFO Studies.
[4] Coase, R. H. (1960), "The problem of social cost", The Journal of Law and Economics, 3, 1-44.
[5] Dutta, B., S. Ghosal and D. Ray (2005), “Farsighted network formation”, Journal of
Economic Theory, 122, 143-164, 2005.
[6] Foley, D. K. (1970), "Lindahl’s solution and the core of an economy with public
goods", Econometrica, 38, 66-72.
[7] Gupta, S., D. A. Tirpak, N. Burger, J. Gupta, N. Höhne, A. I. Boncheva, G. M.
Kanoan, C. Kolstad, J. A. Kruger, A. Michaelowa, S. Murase, J. Pershing, T. Saijo,
A. Sari, 2007: Policies, Instruments and Co-operative Arrangements. In Climate
Change 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [B. Metz, O.R.
Davidson, P.R. Bosch, R. Dave, L.A. Meyer (eds)], Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA.
[8] Harsanyi, J. C. (1977), "Rule Utilitarianism and Decision Theory", Erkenntnis, 11,
pp. 25-53.
[9] Heal, G. (1993), "Formation of International Environmental Agreements", in C. Carraro (ed.), Trade, Innovation and Environment. Dordrecht:Kluwer.
[10] Human Development Report 2008: Fighting Climate Change- Human Solidarity in
a Divided World, UNDP, Palgrave Macmillan.
[11] IPCC, 2007: Climate Change 2007: The Physical Science Basis. Contribution of
Working Group I to the Fourth Assessment Report of the Intergovernmental Panel
on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B.
Averyt, M.Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge,
United Kingdom and New York, NY, USA.
[12] Olson, M., (1971). The Logic of Collective Action: Public Goods and the Theory of
Groups, Harvard University Press, 2nd edition.
[13] Olsen, K. H., (2007), "The clean development mechanism’s contribution", Climatic
Change 84, 59–73.
27
[14] Shapley, L.S. and M. Shubik (1969), "On the core of a system with externalities",
American Economic Review, 59, 678-684.
[15] Starret, D.A. (1973), "Externalities and the core", Econometrica, 41, 179-183.
[16] Stern, N.H. (2007), The Stern Review of the Economics of Climate Change (Cambridge: Cambridge University Press).
[17] UNFCCC. 1997, Kyoto Protocol to UNFCCC. United Nations Framework Convention on Climate Change. Bonn, Germany.
[18] UNFCCC. 2008, Proposal by the G77 & China for A Technology Mechanism under
the UNFCCC.
[19] Wara, M. W. and D. G. Victor (2008), "A realistic policy of international carbon o¤setts", Stanford University Program on Energy and Sustainable Development Working Paper #74.
28
Table 1: Country-wide Measures
Country/Region
EU
China
Type
General emissions reduction/
Renewables
Energy efficiency/ Renewables/
General emissions reduction
Norway
New Zealand
General emissions reduction
General emissions reduction
Costa Rica
Denmark
Reforestation / Emissions
reduction
Renewables
UK
India
Emissions reduction
Energy sector reform
Brazil
Renewables/ alternate fuels
Ecuador
Forest preservation
Spain
Halting desertification
Australia
Halting desertification
Maldives
Adaptation measure
Actions/Goals
20% emissions reduction by 2020 / 20% of region
powered by renewables by 2020
20% improvement in energy efficiency relative to
trend / 20% of region powered by renewables /
reduce emissions intensity per unit of GDP by 20%
- all goals to be achieved by 2020
Zero carbon economy by 2050
10-20% below 1990 levels by 2020, specific plan
outlined for achieving this
No set goal, 1986-2006 added +30% forest cover
to the country / zero carbon economy by 2021
user and prominent exporter of wind power
turbines to the rest of the world
4 new nuclear power stations being built
Comprehensive efficiency and renewables (and
nuclear) energy plan, attempting to create ‘green’
development path
7000 MW production goal for renewables in 2010
/ substantial ethanol production increase
Rainforest sitting on top of oil field protected due
to legislature
4 year program to plant 45 million trees near
desert edge underway
Bio-engineering salt resistant plants and building
desalination plants to combat desertification
Sovereign wealth fund created to purchase new
home should country be permanently flooded
Table 2: Interstate Measures
Regions/Provinces/ States (country)
Rural communities in Southern
Provinces (Mexico)
Type
Watershed Renewal/
Preservation
Iowa, Illinois, Kansas, Michigan,
Minnesota, Wisconsin (US) and
Manitoba (Canada)
General emissions reduction/
cap and trade system
Connecticut, New Hampshire,
Delaware, New Jersey, Maine, New
York, Maryland, Rhode Island,
Massachusetts and Vermont (US)
General emissions reduction/
cap and trade system
Arizona, California, Montana, New
Mexico, Oregon, Utah and Washington
(US), British Columbia, Manitoba,
Ontario and Quebec (Canada)
General emissions reduction/
cap and trade system
Western Governors Association
member states (US)
renewables
Actions/Goals
Payment to farmers for preserving and
maintaining forest area deemed ecologically
critical (Up to roughly 40 USD/ hectare/ year)
Midwest greenhouse gas reduction accord –
works similarly to the EU model for emissions
reduction/ trading, only on a shorter 30 month
timescale – mandates group and individual
reduction targets for this time
Regional Greenhouse Gas Initiative – cap power
sector emissions at 188 million tons
immediately, tightened by 10% by 2018 –
member policing of each other – pure auctioning
ETS, with offsets allowed
Western Climate initiative – cap and trade
approach modeled in phases as in the EU, focus
on all institutes emitting over 25,000 tons of
CO2 annually, target is 15% emissions reduction
from base year 2005 by 2020, enforced on a firm
by firm basis rather than standard state basis.
Clean and Diversified Energy Initiative – 30000
MW capacity goal in renewables for the group
by 2015, 20% group energy efficiency
improvement goal by 2020
Table 3: Individual State Measures
Province/ State (country)
Almost all Provinces
(Canada) and States (US)
Type
General emissions
reduction/ energy efficiency
California (US)
Various – focus on
renewables and integration
of them into infrastructure
Florida (US)
Various – focus on
decentralizing power grid
through renewables
British Columbia (Canada)
Carbon tax
New South Wales
(Australia)
Energy efficiency / adaption
preparedness
Queensland (Australia)
Renewables & efficiency
R&D
Karnataka (India)
renewables
Actions/ Goals
Goals vary by state/province but most have a
minimum of at least 10% improvement in energy
efficiency and emissions reduction by various
means by 2020
Samples - Global Warming Solutions Act of 2006,
Alternative and Renewable Fuel and Vehicle
Technology Program, Solar Water Heating and
Efficiency Act of 2007, The Clean Car Law, and
an 80% GHG reduction target by 2050 from base
year 1990
Samples – 40% reduction target from base year
1990 by 2025, 80% by 2050, creation of Florida
Climate and Energy Commission to oversee
policy, banning of common energy saving
devices prohibited, Florida Climate Protection
Act, tax break for residential property with
installed renewable power generators, Florida
Energy Systems Consortium created to
encourage R&D
Revenue neutral tax easing in from a $10 CAD
carbon price equivalent tax in 2008 to a $30 one
in 2012, specific goods taxed by average carbon
content
$63 million (AUD) program to help 220 000 low
income households incorporate new technology
into the home for improved energy and water
efficiency
$7 million (AUD) into various projects aimed at
developing renewable technology and upgrading
energy efficency
A ‘100 000 solar roofs’ project centered around,
Amazonas (Brazil)
Rainforest preservation
but not limited to, the Bangalore area. To be
achieved by 2013.
Moved in opposition to Federal ethanol
production mandates, halted practice of handing
out chainsaws to ‘promote development’,
adopted clear preservationist stance
Table 4: City-Level Measures
City
Copenhagen (Denmark)
New York City (US)
Type
Emissions & pollution
reduction
Adaptation measures /
emissions reduction/
energy efficiency
Toronto (Canada)
Energy efficiency/
renewables
Munich (Germany)
Energy efficiency/
renewables / financial
innovation
London (UK)
Renewables/ energy
efficiency
Energy efficiency/
emissions reduction /
green industry
promotion
Renewables/ energy
efficiency / emissions
reduction
Shanghai (China)
Sydney (Australia)
Actions/Goals
City wide public bike system
Major infrastructure project – estimated 30%
reduction in emissions. Project includes improving
city forest cover (erosion concern), reworking of
electrical grid and building codes for energy
efficiency, replacing all aging water pipes with focus
on durability, and mandates on increased minimum
allowable fuel efficiencies for cars
Infrastructure upgrading for durability, updating
building codes for energy efficiency, green roof
promotion, large number of wind power generators
added, both on land and over water
Infrastructure and building code upgrades (similar to
NYC and Toronto), wind power generators added,
weather based derivatives designed to manage wind
power financial risks of variable winds
Major infrastructure and building code upgrades for
energy efficiency, various wind and solar projects
80 billion Yuan for tree & foliage planting & water
infrastructure upgrading, revamped tax structure to
encourage green industry, plan for removal of 65% of
vehicle traffic in the city
Carbon neutral goal by 2020, energy efficiency and
water infrastructure upgrades, building code
revamped for energy efficiency and carbon neutral
goal